We're a small team that deeply feels the risks from advanced AI. We're committed to conducting open science, but we're increasingly concerned about whether external safety organisations like ours will continue to have access to the resources we need.
Give our write-up a read!
There's a ๐ ๐ซ๐จ๐ฐ๐ข๐ง๐ ๐ซ๐๐ฌ๐จ๐ฎ๐ซ๐๐ ๐ ๐๐ฉ ๐๐๐ญ๐ฐ๐๐๐ง ๐ฌ๐๐๐๐ญ๐ฒ ๐ญ๐๐๐ฆ๐ฌ ๐ข๐ง๐ฌ๐ข๐๐ ๐๐ซ๐จ๐ง๐ญ๐ข๐๐ซ ๐ฅ๐๐๐ฌ ๐๐ง๐ ๐ข๐ง๐๐๐ฉ๐๐ง๐๐๐ง๐ญ ๐ฌ๐๐๐๐ญ๐ฒ ๐ซ๐๐ฌ๐๐๐ซ๐๐ก๐๐ซ๐ฌ, and closing it is essential to maintaining an ecosystem of impactful safety research.
We've spent months securing a large, multi-year GPU contract. This taught us that ๐๐ฎ๐ง๐๐ข๐ง๐ ๐ข๐ฌ ๐ง๐๐๐๐ฌ๐ฌ๐๐ซ๐ฒ ๐๐ฎ๐ญ ๐ง๐จ๐ญ ๐ฌ๐ฎ๐๐๐ข๐๐ข๐๐ง๐ญ. We've written up the non-obvious bottlenecks we hit, and what can be done to alleviate them: https://t.co/gaqtAd4g5l